TIG Welding Aluminum Tips: Settings, Technique and Fixes
Table of Contents
A weekend fabricator lays down what looks like a perfect stack of dimes on 6061 plate, then the part cracks overnight and porosity shows up under the bead. Almost every aluminum TIG failure traces back to a handful of controllable details, and mastering them is what this guide walks through.
Why Aluminum TIG Welding Punishes Small Mistakes
A weekend fabricator tacked up two pieces of 6061-T6 plate, ran what looked like a textbook stack of dimes, and went in for the night. By the next afternoon the part had cracked along the toe, and a quick etch revealed porosity sitting right under that pretty surface. Nothing in the bead’s appearance warned him. That is the core frustration of aluminum TIG: the metal happily accepts a weld that looks perfect, then rejects it structurally hours later.
So what actually makes TIG welding aluminum harder than steel? It is not the melting point — aluminum melts at roughly half the temperature of mild steel. The real difficulty is that four of aluminum’s physical habits work against you at the same time: a hard oxide crust that will not melt with the base metal, heat that runs away three to five times faster than in steel, expansion and shrinkage that distort and crack joints, and zero color change right up to the moment the puddle collapses.
This article is written as a practical, tip-by-tip manual: machine settings, tungsten and gas setup, cleaning, filler selection, hand technique, defect fixes, safety, and a practice plan. Everything below starts with the physics, because once you understand why aluminum misbehaves, the individual tips stop feeling like arbitrary rules and start making sense.
The Oxide Skin That Will Not Melt With the Metal
The instant bare aluminum meets air, it grows a skin of aluminum oxide — thin, invisible, and extremely stubborn. Portland Community College’s WLD 222 course material puts the number plainly: aluminum oxide melts at approximately 3,300°F, while the base metal underneath gives way somewhere around 1,100–1,200°F depending on alloy content (PCC WLD 222 course material). That is a crust roughly three times hotter than the metal it sits on.
Steel does not behave this way. Mild steel’s mill scale softens and dissolves into the weld pool as you go, so it never blocks the arc. Aluminum’s oxide just sits there and insulates the puddle, which is why contamination and poor prep show up as lack of fusion, porosity, and wandering arcs far more often on aluminum than on steel. It is also why alternating current is non-negotiable: the cleaning half of the AC cycle exists specifically to blast that crust away.
Two practical consequences follow. First, cleanliness is a setting, not a courtesy — a meticulously ground tungsten and a perfect AC balance will not save an oily joint. Second, aluminum forgives nothing, so the Aluminum Cleaner for Welding prep sequence covered later in this article is what keeps hydrated oxide and workshop grime out of your puddle in the first place.
Heat Runs Away Three to Five Times Faster Than in Steel
Aluminum conducts heat roughly three to five times faster than mild steel. Put a puddle on steel and the surrounding metal warms and stays warm; put the same arc on aluminum and the plate acts like a heat sink, pulling energy away as fast as the machine can deliver it.
The result is counterintuitive for anyone who learned on steel: you usually need more current than steel intuition suggests, despite aluminum’s much lower melting point. The metal takes current sluggishly at first, then floods with heat all at once near the end of the joint. That late flood is why craters and terminal cracks are so common, and why Miller’s beginner guidance emphasizes foot-pedal control and downslope rather than fixed amperage. Thicker sections demand proportionally more current, travel speed has to stay brisk, and long beads on thin material need heat sinking or stitch techniques. The same fact explains why Aluminum MIG welding fights identical heat-sinking with spray transfer and push technique instead of a pedal.
It Distorts, Shrinks and Cracks as It Freezes
Aluminum’s coefficient of thermal expansion is roughly double that of steel, and the weld metal itself shrinks about 6% in volume as it solidifies — roughly twice what you would budget for on steel. Add a joint that heats unevenly because of conductivity, and distortion is not a possibility, it is a certainty unless you plan for it. That combination cracks restrained joints overnight, pulls fillet toes, and wrecks asymmetric assemblies. It is the same physics whether you are stitching a marine alloy wheel repair or running a long seam on a boat hull: stagger tacks, balance heat side to side, keep beads short, and finish with proper downslope and crater fill rather than simply releasing the pedal.
Quick Note: If your part looks straight when it comes off the fixture but sits crooked the next morning, shrinkage, not your arc, is the culprit. Fixture hard, tack more than feels necessary, and weld in alternating segments so pull forces cancel out.
No Color Change Until It Is Too Late
Steel glows. It goes straw, then purple, then cherry red, giving the welder a continuous visual temperature gauge built into the arc. Aluminum gives you nothing. It stays the same silver color right up until it suddenly goes shiny and wet, then collapses into a dark crater with oxide floating on top. Waiting for color is waiting for a signal that never comes. You have to read the puddle instead: a healthy aluminum puddle looks like a small, bright mirror with a fine ripple, and the base metal does not tint around it. Push it a hair too far and it slumps and loses definition. Puddle reading rather than color guessing is what separates welders who control aluminum from welders who react to it.
| Property | Mild steel | Aluminum (6061) | What it means at the bench |
|---|---|---|---|
| Melting point | Roughly 2,500°F and up | About 1,100–1,200°F | Less heat to melt, far less margin once molten |
| Oxide behavior | Scale softens into the puddle | Oxide melts near 3,300°F | Cleaning and AC cleaning action are mandatory |
| Thermal conductivity | Baseline | 3–5× faster | Higher current, faster travel, sudden heat buildup |
| Thermal expansion | Baseline | About 2× steel | More pull, more distortion, more warping |
| Solidification shrinkage | Roughly half of aluminum | About 6% by volume | Craters, toe cracking, joint movement on cooling |
| Temperature indicator | Visible glow before melting | No color change | Read the puddle, not the heat tint |
Read that table top to bottom and the theme is the same: aluminum removes every crutch a welder gets on steel and replaces it with tighter tolerances for preparation, settings, and hand skill.
Here is the route map. First we look at why alternating current exists and how its cleaning action strips oxide. Then equipment: torch, tungsten, and gas flow that keep the arc stable. After that, concrete starting settings for amperage, AC balance, and AC frequency; the cleaning and joint prep workflow no machine can compensate for; filler selection between the 4043 and 5356 families; the hand skills of angle, arc length, and filler timing; a symptom-to-cause-to-fix troubleshooting table; the safety and ventilation details that matter specifically for aluminum TIG; and finally a progressive practice plan plus a shop-floor checklist.
Before you touch a single amperage setting, though, there is one piece of physics worth understanding properly: why the current has to swing back and forth instead of running in one direction. That is where we start.
Why AC Is Used and How the Cleaning Action Works
Ask ten welders why aluminum is welded on AC and you will get answers ranging from “that is just the setting” to “something about the oxide.” The real answer is physics, and once you understand it, most of the machine knobs on a TIG welder stop being mysterious. Aluminum and magnesium carry a stubborn oxide skin that melts around 3,300°F while the metal underneath melts near 1,100–1,200°F. If that oxide is not actively removed during welding, it floats into the puddle and creates inclusions, dirty beads and lack of fusion. Alternating current is the only practical way to melt base metal and strip oxide in the same pass.
What Each Polarity Actually Does
Gas tungsten arc welding gives you three choices of current, and each one behaves differently on aluminum.
| Polarity | Where the heat goes | Oxide cleaning | Result on aluminum |
|---|---|---|---|
| DC electrode negative (DCEN) | About 70% at the workpiece | None | Deep penetration, oxide left intact, dull contaminated bead |
| DC electrode positive (DCEP) | About 70% at the electrode | Strong | Clean puddle but tungsten overheats, erodes and softens |
| AC (alternating) | Roughly 50/50 split | Present on the positive half cycle | Penetration plus cleaning in one arc |
DCEN is what you would use on steel or stainless: a tight, hot, penetrating arc with a cool electrode. On aluminum it gives you beautiful penetration and a filthy puddle, because nothing ever lifts the oxide film. Flip to DCEP and the cleaning is excellent, but roughly seven-tenths of the arc’s energy lands on a tungsten that is already near its limit. The electrode balls up, the arc goes erratic and tungsten eventually spits into the weld. That is why DCEP alone is reserved for very short, low-amperage cleanup work rather than production welding. As AWS’s overview of the GTA process explains, alternating current is specified for aluminum and magnesium precisely because it alternates between the two jobs.
How the Oxide Layer Gets Blasted Off
The cleaning half cycle does not work by melting anything. Aluminum oxide behaves like a tiny capacitor: as the electrode goes positive, ions accumulate across the oxide film and build an electrical charge. When the voltage across that layer reaches its breakdown point, the oxide ruptures and is physically ejected from the surface as fine particles. This happens well below the melting point of the base metal, which is why the etched “cleaning zone” surrounding an AC bead looks frosted rather than melted.
The width of that frosted band is your visual feedback on balance. More time on electrode negative means a narrower cleaning zone and deeper penetration; more time on electrode positive widens the etch and flattens the bead. The practical implications are covered in detail in Miller’s AC balance control guidance, and the same reasoning applies whether you are welding plate, extrusion or a repair on a cast housing. If you are new to aluminum prep altogether, our aluminum cleaner for welding guide covers the manual side of the same problem.
Why High Frequency Is Non-Negotiable on Transformer Machines
AC current crosses zero sixty times a second on a 60 Hz supply, and at each crossing the arc momentarily dies. On a classic transformer machine, a high-frequency signal in the tens of kilohertz is superimposed onto the AC wave to re-strike the arc before it can extinguish, giving a smooth, continuous arc instead of a stuttering one. On modern inverter welders the AC waveform is generated electronically and the arc is far more stable on its own, but the high-frequency start circuit still lets you strike the arc without touching the tungsten to the work. Touch-starting on aluminum is a fast route to contamination, so keep the HF doing its job.
Quick Note: Cleaning action also depends on shielding gas quality. AWS A5.32 specifies GTAW argon at a minimum purity of 99.997% with a maximum dew point of −60°C (−76°F). Wet or contaminated gas shows up as porous, cloudy beads no matter how good your technique is.
That gas spec matters more than it sounds. A marginal argon supply steals cleaning efficiency and invites porosity, which is one reason aluminum MIG welding setups are so particular about flow rates and hose condition too. When troubleshooting starts with “the arc looks weak or the bead is dirty,” check gas before you touch the balance knob. Flux-based processes exist as an alternative when gas coverage is impractical, and our look at aluminum flux welding explains where that trade-off makes sense.
Bottom line: DCEN gives you penetration, DCEP gives you cleaning, and AC gives you both in the ratio you choose. Everything in the next section, from balance percentages to frequency settings, is just a way of steering that split.
Torch, Tungsten and Gas Setup That Sets You Up to Win
Most aluminum TIG problems start before the arc does. Undersized torch, small cup, wrong tungsten alloy, wrong tip shape, and no amount of foot-pedal finesse saves the weld. Getting this hardware layer right is the cheapest speed-up in the whole process.
Pick a Torch That Can Take the Heat
Aluminum loves current, and current means heat in the torch body. For bench work on material up to about 1/4 inch, a small air-cooled torch in the 150-amp class is comfortable and handy. Beyond that, or on anything you’ll run at high duty cycle, go water-cooled and match the torch rating to your machine’s output at its actual duty cycle, not the peak number on the spec sheet. A torch rated for 200 amps at 60 percent duty cycle will overheat fast if you lean on 200 amps continuously without coolant flow.
Cup size matters more than people expect. A #6 or #7 cup is a fine general choice, but stepping up to a #8 or #12, and better yet fitting a gas lens, gives a wider, calmer column of argon that tolerates a slightly longer arc and a slightly shakier hand. A gas lens lets you drop flow rates a little while improving coverage, which is a real advantage on a material that oxidizes the instant it’s exposed to air.
Argon Flow: 15–20 CFH and No Higher
For manual TIG on aluminum, 15–20 cubic feet per hour of argon is the standard starting window. The instinct to crank flow up “to be safe” backfires: excessive velocity aspirates turbulent air into the shielding envelope, and the result is porosity, not protection. If you’re using a large cup or gas lens in a drafty shop, confirm coverage with a soap-bubble leak check and a quick test weld rather than guessing. A quality AllGas welding supply cylinder setup with a proper regulator makes consistency far easier than a cheap flowmeter that drifts.
Quick Note: More flow is not more shielding. If your porosity disappears when you shield the draft instead of opening the valve, the valve was never the problem.
Choosing the Tungsten
This is where polarity really bites. AC puts substantially more heat into the electrode than straight electrode-negative work, so the same amperage needs a larger-diameter tungsten on AC than it would on DCEN. A 3/32-inch electrode that handles 100 amps happily on steel can glow cherry red and erode on aluminum at that current. Sizes roughly scale like this:
- 1/16 in (1.6 mm): up to about 60–80 amps, thin sheet
- 3/32 in (2.4 mm): about 80–160 amps, the everyday workhorse
- 1/8 in (3.2 mm): 160–250 amps and up
For alloy type, modern inverter machines respond best to ceriated (gray band), lanthanated (blue band) or zirconiated (brown band) electrodes. These hold a tip, start easily and tolerate AC well. Pure green-band tungsten is not recommended for inverters — it was designed for old transformer AC arcs, balling up under their higher voltage, and on an inverter it tends to erode, spit and wander. If you’re still running a transformer machine, pure or zirconiated is the traditional choice and works fine there.
Avoid 2 percent thoriated (red band) on AC if you can. It runs, but it erodes and oxidizes badly under alternating current, and it carries a low-level radioactive grinding-dust hazard that’s simply unnecessary when lanthanated does the same job. For help reading electrode markings, the Fabricator’s guide to tungsten type and size for aluminum GTAW is a solid reference, and AWS defines the electrode classification and shielding practice in its GTAW overview.
Grind It Right Every Time
Grind at a 15–30 degree included angle and grind lengthwise, straight down the length of the electrode, never around the circumference. Radial grinding leaves circular scratches; the arc follows those grooves and circles the tip, which is exactly the “arc wander” you’re trying to avoid. Longitudinal scratches give the arc a straight runway to the tip. Use a dedicated diamond wheel, keep the electrode perpendicular to the wheel, and never contaminate it with the wheel you use for steel.
Tip shape: on modern inverter AC, a lightly balled or slightly blunted point outperforms a needle-sharp one, which erodes within seconds. On transformer AC you want a proper ball tip, formed by starting the arc on a scrap plate and letting current round it over. Tungsten that’s contaminated from touching the puddle must be reground or snapped back; you cannot weld aluminum with a tungsten that has pickup on it.
Stickout should stay short: about one cup diameter or less. Long stickout loses shielding and invites wander, so keep the electrode close to the work, and if you need to see into a tight joint, change cups rather than extending the tungsten. These setup habits carry straight over to other aluminum processes — the same logic on prep and shielding appears in our aluminum MIG welding guide.
Dialing In Amperage, AC Balance and AC Frequency
Once your torch, tungsten and gas are sorted, three numbers on the machine decide whether the puddle wets in cleanly or fights you all the way to the end of the joint: amperage, AC balance and AC frequency. Treat them as a connected set rather than three independent dials, because turning one changes what the other two need.
Amperage: Start With the Thickness Rule, Finish With Your Foot
A dependable starting point for aluminum is the familiar thickness rule: roughly one amp per 0.010 in (0.25 mm) of material. So 1/16 in plate starts near 60–65 A, 1/8 in plate near 110–125 A, and 3/16 in near 150–180 A. Miller’s beginner TIG aluminum walkthrough walks through exactly this kind of thickness-based chart, and it is the right place to begin before you change anything else.
But that number is a ceiling, not a target. Aluminum pulls heat away from the joint three to five times faster than steel, so the puddle comes up slowly at the start of a cold bead. Once the surrounding metal saturates, the same amperage that felt lazy suddenly becomes too much. That is why the foot pedal matters more on aluminum than on almost any other metal:
- Strike the arc and hold around 65–75% of pedal travel until a puddle slightly wider than your filler rod appears.
- Back off slightly as the joint heats up, then modulate between the two as you add filler.
- Creep the pedal down in the last half inch of a butt joint, where trapped heat is highest.
| Material | Starting amperage | Starting balance | Starting frequency |
|---|---|---|---|
| 1/16 in (1.6 mm) sheet | 60–80 A | 75% EN | 120–140 Hz |
| 1/8 in (3.2 mm) plate | 110–125 A | 75% EN | 110–120 Hz |
| 3/16 in (4.8 mm) plate | 150–180 A | 75% EN | 100–110 Hz |
| Castings, heavy buildup | By section | 65–70% EN | 60–90 Hz |
AC Balance: Percent Electrode Negative
AC balance is simply how much of each cycle is spent on electrode negative (penetration) versus electrode positive (cleaning). It is displayed as percent EN on most inverters, and it is the single most misunderstood setting on an aluminum TIG machine. Miller’s AC balance control guide explains the trade-off clearly: more EN drives the arc deeper and narrows the bead but etches less of the oxide away; less EN widens and flattens the puddle with a stronger cleaning zone around it.
The practical ranges look like this:
- 75% EN — the everyday starting point for clean, fit-up-tight 6061 and 5356 work.
- 65–70% EN — heavily oxidized, mill-scaled or slightly dirty material that needs more cleaning action.
- Up to 80% EN — very clean, precisely fit joints where you want deeper penetration and a narrower, flatter bead.
Quick Note: Black peppering or dark specks appearing in the puddle mean the cleaning half of the cycle is too weak. Drop the EN value in 5% steps. Roughly 60% EN is the practical floor — if you still see peppering there, the machine is not the problem, and proper mechanical cleaning before you strike the arc is the real fix.
AC Frequency: Shaping the Arc
Frequency does not change the balance of heat; it changes the shape and focus of the arc cone. The everyday sweet spot is 100–120 Hz, and 120 Hz is the common factory Pro-Set default on many inverters, with an adjustable range of roughly 60–150 Hz. Raise it and the arc column tightens, stiffens and stands more upright, which is exactly what you want on thin material, overhead positions and inside fillet joints where a wandering wide arc will undercut the toes. Drop it and the arc softens and spreads, which suits castings, porous or dirty surfaces, and any situation where you are building up a worn area rather than running a precise bead. The Fabricator’s overview of aluminum machine parameters makes the same point: frequency is an arc-shaping tool, not a penetration tool.
Expect the arc to get audibly louder and higher-pitched as you push frequency up. That squeal is normal, not a sign of a fault, and it is worth mentioning to anyone working nearby.
Pulse Settings for Thin and Out-of-Position Work
Pulsing gives the puddle time to solidify between peaks, cutting total heat input and letting you weld thin sheet or vertical joints without the puddle running away. A typical starting point on an inverter is 100 pulses per second, 40% peak time, and background amperage set around 25% of your peak value. From there, lower the background toward 15–20% if the puddle still grows too fast, and slow the pulse rate if you cannot keep a rhythm between dabbing and travel. Pulse frequency and AC frequency are separate controls — one sets your dabbing rhythm, the other sets arc focus.
Short Note: Change one variable at a time. Dial in amperage on a scrap bead first, then adjust balance for cleanliness, then frequency for arc shape. Chasing all three at once leaves you with no idea which change actually improved the weld.
Cleaning and Joint Prep: The Step Everyone Skips
Ask a hundred welders why an aluminum job failed and you will hear answers about amperage, tungsten type, AC balance and pedal control. Almost nobody volunteers that the plate was dirty. Yet in shop after shop, a leading cause of rejected aluminum welds is not a machine setting at all. It is a joint surface that never got properly cleaned. No dial position can compensate for oil, coolant residue, moisture or a layer of hydrated oxide clinging to the joint. If you change one habit after reading this article, make it this one: treat cleaning as part of the welding process, not as busywork you rush through.
Why Hydrogen Is the Real Enemy
Here is the fact that reframes everything: all porosity in aluminum welds is caused by hydrogen. Not most of it, all of it. Hydrogen is soluble in molten aluminum and nearly insoluble in solid aluminum, so as the puddle solidifies the dissolved gas is squeezed out and trapped as bubbles. Those bubbles are your porosity, your pinholes and your worm tracks. The TWI engineering body identifies the sources of that hydrogen as hydrocarbons such as oil, grease, cutting fluids and marker pen ink, moisture from humid air or wet material, and hydrated oxide formed when bare aluminum sits in a damp environment.
That last one catches people out. The oxide layer on aluminum is unavoidable, but freshly exposed aluminum immediately begins chemisorbing water from the atmosphere and forming aluminum oxide hydroxide. That hydrated layer holds water tenaciously, and it will happily release hydrogen straight into your puddle while it breaks down in the arc. Lincoln Electric’s process guidance on TIG welding aluminum puts cleanliness first for exactly this reason.
Quick Note: Porosity you can see on the surface is the polite version. The dangerous case is a bead that looks like a perfect stack of dimes while hydrogen bubbles hide under the crown. That is why cleaning discipline matters even when the weld “looks fine.”
The Order Matters: Solvent First, Brush Second
The most common prep mistake is using the right tools in the wrong sequence. If you wire brush an oily or greasy plate, the brush drives the hydrocarbon contamination deeper into the softened surface instead of removing it. You end up welding over a plate that looks scuffed and clean but is smeared with embedded oil.
The correct sequence, as laid out in Miller’s beginner TIG aluminum procedure, is simple:
- Degrease first. Wipe the joint area with a clean, lint-free cloth and acetone or a dedicated aluminum prep solvent. Never re-use a rag once it is visibly dirty. For heavier contamination, flood the joint with solvent and wipe in one direction so residues leave the surface instead of moving around.
- Brush second. Only once the surface is dry and oil-free should you hit it with a stainless steel wire brush. Use dedicated, aluminum-only tools, because a brush previously used on steel leaves embedded iron particles that discolor and corrode the weld.
- Brush in one direction with short, overlapping strokes. Back-and-forth scrubbing simply redistributes the oxide and debris you were trying to remove.
- Inspect and re-clean. A properly prepared surface has a uniform dull silver sheen with visible directional scratches. If you can still see a shine, there is oxide left.
For solvent choices, commercial prep products and household substitutions that do and do not work, our aluminum cleaner for welding guide covers the options in detail.
Cut Edges, Castings and Contaminated Zones
Mill-finish extrusion and plate are the easy cases. Trouble starts with sheared edges, plasma-cut profiles, castings and anything that has seen service. Sheared and cut edges often carry a smeared lip of mechanically displaced metal that will not wet out properly and traps oxide under the bead, producing a classic lack-of-fusion defect. Grind or file the smear away until you reach clean parent metal, and blend back any burnt or oxide-heavy zone from thermal cutting. Cast parts such as engine components and wheels need even more aggressive attention because they hold oil in their porosity; our alloy wheel welding guide covers that workflow, including the preheat and bake-out steps porous castings demand.
Storage, Acclimation and the Eight-Hour Window
Aluminum’s high thermal conductivity also makes it a condenser. Sheets stacked flat in a humid shop trap moisture between the layers, and that water sits directly against fresh metal, forming hydrated oxide right where you intend to weld. Store stock dry and standing vertically, separate stacked sheets with stickers or spacers, and give material time to acclimate in the welding area. Most training programs recommend letting stock sit for about 24 hours so it reaches ambient shop temperature before you strike an arc. The same logic applies to parts coming out of cold storage or a refrigerated truck.
Time after cleaning matters just as much. Freshly brushed aluminum begins re-oxidizing and re-hydrating immediately, and course guidance commonly advises welding within eight hours of final cleaning. Beyond that window, give the joint another solvent wipe and a fresh brush pass. If the part sat overnight in a damp shop, assume it is contaminated again regardless of how carefully it was prepped.
The Workflow to Make Automatic
Turn cleaning into a repeatable sequence rather than a decision you make each time: degrease, brush, inspect, weld. Heavy, established oxide cannot be brushed away reliably, so where preheat is acceptable, gently warming the joint ahead of welding helps drive off bound moisture. Keep filler rod as clean as the parent metal, since a fingerprint on 1/8-inch rod deposits exactly as much contamination as one on the plate.
Note: If you are welding marine or coastal material, salt contamination is a separate hazard on top of ordinary shop grime. Wash and dry the part thoroughly before you ever bring it to the table; our aluminum boat welding guide explains why.
Bottom line: cleanliness beats every machine setting on this list. You can dial in perfect amperage, balance and frequency and still destroy a part with a contaminated joint. Get the surface right, and the settings in the next section will do exactly what you expect them to.
Choosing the Right Filler: 4043 vs 5356 and Beyond
You can dial in perfect AC balance, run a clean tungsten, and still ruin a joint by feeding the wrong rod. Filler metal is the one variable that changes crack sensitivity, strength, anodized color, and how the part behaves years down the road. The good news is that the choice rarely gets complicated: two alloys handle the overwhelming majority of work. Industry welding guides and AWS C5.5 recommended practices for aluminum welding both point out that ER4043 (and its newer sibling ER4943) and ER5356 account for well over 85% of all aluminum weldments made. Learn these two well and you can weld almost anything a home shop or job shop throws at you.
Start With the Base Metal, Not the Rod Box
Filler selection works best as a short decision tree rather than a guess. Ask these questions in order:
- What series is the base metal? 6xxx (6061, 6063) and cast alloys are forgiving territory for 4043. 3xxx and 5xxx alloys lean toward 5356.
- Does the base metal contain more than 2.5% magnesium? If yes, use 5356. Do not use 4043 or 4943 on these alloys, because the high-magnesium base and the silicon filler form a coarse, brittle intermetallic zone in the fusion boundary.
- Will the finished part see sustained service above 150°F (66°C)? If yes, never use 5356. Switch to 5554 or 4043/4943.
- Is the part anodized after welding? Anodized appearance decides more jobs than people expect.
- Static load or dynamic load? Cyclic, impact, or marine structural service favors 5356 for ductility.
Head-to-Head: ER4043/4943 vs ER5356
| Property | ER4043 / ER4943 | ER5356 |
|---|---|---|
| Main alloying element | Silicon (~5%) | Magnesium (~5%) |
| Crack and distortion resistance | Excellent; very low shrinkage cracking | Good, but more crack-prone on 6xxx |
| Wetting and flow | Smooth, fluid, forgiving puddle | Stiffer puddle, narrower wetting |
| Shear strength of fillet welds | About three fillet passes to equal one 5356 pass | Higher single-pass strength |
| Ductility | Moderate | High |
| Anodized appearance | Dark gray to black bead | Close to base metal, whitish |
| Sustained service temperature | Suitable for elevated temperature | Avoid above 150°F sustained |
| Typical work | 6061, castings, thin sheet, decorative | 5xxx, marine, anodized, structural |
When 4043 and 4943 Win
Silicon fillers melt and flow like butter. That fluidity is why 4043 is the default for 6061 sheet and plate, for cast aluminum wheel repair, and for anything where a smooth, attractive bead matters. Silicon also suppresses hot cracking, so 4043 tolerates dirty-ish joints and restraint better than 5356. It is also the right choice for elevated-temperature service, which is why 4043 shows up on engine parts, manifolds, and exhaust-adjacent fabrications. ER4943 is the upgrade path: it keeps the crack resistance of 4043 while adding enough magnesium to deliver strength nearer 5356 on 6061-T6 joints, which makes it a smart pick for structural 6xxx work that will not be anodized.
When 5356 Wins
If the base metal is 5052, 5086, 5456, or any 5xxx alloy with more than 2.5% nominal magnesium, 5356 is not optional. It is also the correct rod for anodized parts, because it anodizes to a light, near-base-metal tone instead of the dark gray stripe a 4043 bead leaves behind. Marine fabricators live on 5356 for the same reason boat builders specify it for hull joints: high ductility and excellent resistance to dynamic and cyclic loading. The trade-off is a stiffer puddle, weaker wetting on castings, and a higher tendency to crack on 6xxx under heavy restraint.
Quick Note: The 150°F rule matters more than it sounds. Above that temperature, the magnesium-rich 5356 joint can suffer progressive strength loss over time. For exhaust-adjacent, engine, or process piping work, reach for 5554 (or 4043/4943) instead, even if the base metal is a 5xxx alloy.
Rod Diameter and Keeping Rod Clean
Match filler diameter to the thinnest member you are joining: roughly 0.035 in (0.9 mm) for sheet under 1/16 in, 1/16 in (1.6 mm) for 1/8 in material, and 3/32 in (2.4 mm) for thicker sections. A rod that is slightly smaller than the joint thickness lets you control heat input without drowning the puddle.
Clean rod is as important as clean plate. Filler picks up oil from bare hands and moisture from humid storage, and hydrogen from either source becomes porosity in the bead, as TWI’s porosity guidance explains. Handle rod with clean gloves or a lint-free cloth, keep lengths in a covered tube or dispenser, and wipe with a solvent-rated aluminum cleaner for welding if it has been sitting. Never dip hot rod into anything, and never let it rest on a welding table covered in grinding dust.
That covers the consumables. Next, the hand skills that make all of these settings actually show up in the bead.
Hand Skills: Torch Angle, Arc Length, Pedal and Filler Timing
Settings get you in the game. Hands win it. Dial in perfect AC balance and you can still lay a lumpy, undercut bead if your torch wanders, your arc creeps long, or your rod stabs the tungsten. Aluminum gives no color warning before the puddle collapses.
Brace First, Then Weld
Before you strike an arc, find a brace. Rest your torch hand, forearm, or both on the table, a clamp, or scrap so only your wrist and fingers move. Freehand works on short welds; over a 12-inch bead it guarantees arc drift and inconsistent penetration. Brace the filler hand too, with the rod resting across your fingers or propped on a guide. On long seams, reposition your body instead of reaching past your comfortable range.
Torch Angle and Filler Angle
Two angles matter, and they do different jobs. The torch tilt runs about 5 to 15 degrees back from vertical, trailing the direction of travel. That keeps the cup from blocking your view and pushes molten metal forward instead of piling it behind you.
The filler rod enters from the front at roughly 10 to 15 degrees from perpendicular. Feed at a steep angle and you disturb the puddle, drag oxide into the weld, and risk touching the tungsten. Keep it low, with the rod tip staying at the puddle’s leading edge.
Quick Note: Consistency beats perfection. Pick one torch angle and one filler angle and practice until both are automatic. Changing angles mid-bead is what creates the wandering ripple pattern beginners hate.
Arc Length: The Biggest Variable
Keep the tungsten-to-work distance about equal to the electrode diameter, and never beyond 1/4 inch. A 3/32-inch tungsten means a 3/32-inch arc. That single number controls heat input, bead width, and how far the arc spreads.
Too long and the arc fans out, heat scatters, the puddle runs wide and shallow, and penetration drops. Too short and you dip the tungsten, forcing a stop, regrind, and restart. Aluminum pulls heat away from the joint three to five times faster than steel, so a tight arc keeps energy where it belongs.
Each time you dip, break off the contaminated tip and regrind. A fouled electrode gives an erratic arc, black peppering, and inconsistent penetration, and invites arc wander, that sideways drift that leaves you chasing the seam. Regrind with straight longitudinal scratches running up the tungsten, never swirling or crosswise marks that disrupt the arc. For electrode choice and prep, see prepping aluminum the right way.
Pedal Control and Filler Timing
The foot pedal is your throttle, and on aluminum you work it constantly. Start with a fast burst of current to form the puddle, ease off to a comfortable amperage as you travel, then taper down gradually at the end rather than snapping off.
Filler timing is rhythm. When the puddle reaches the size and wetness you want, dab the rod at the leading edge, withdraw slightly, move forward, and dab again. That dab momentarily cools the puddle, which aluminum needs because it solidifies fast and your heat is racing away from the joint. Consistent rhythm produces even ripples and even penetration.
Rod diameter matters more than most welders realize. Undersized filler removes the cooling effect of rod entering the puddle and can cause burn-through on thin stock. If your 0.045-inch rod cannot keep up, step up to 1/16 or 3/32 rather than welding faster. Clean rod also prevents porosity, the same wire-handling principle covered in the complete practical aluminum MIG guide.
Travel Speed and Terminating the Weld
Hold a steady pace so the puddle stays the same size down the seam. Speeding up thins the bead, slows penetration, and leaves lack of fusion at the root. Slowing down overheats the puddle, widens the heat-affected zone, and invites distortion, a real problem given aluminum’s roughly double thermal expansion.
The end of the weld is where good beads turn bad: crater cracks usually come from breaking the arc improperly. Instead of releasing the pedal, reverse direction and weld back into the previous bead while tapering current down, then hold post-flow so the hot tungsten and weld stay shielded as they cool. Miller’s advanced aluminum welding guide covers termination in depth.
Practice Drills That Build Skill
Run these on scrap before you touch a real part:
- Bead-on-plate: Straight beads on 1/8-inch 6061 at fixed settings, focusing only on arc length and travel speed.
- Dab rhythm: Add filler on a metronome count and aim for even ripple spacing.
- Arc length: Weld watching only tungsten height and restart whenever it drifts above a tungsten diameter.
- Stop-start: Terminate and restart mid-plate to master downslope, post-flow, and tie-in.
- Can-crush test: Weld the base of an aluminum can on low amps to build fine pedal control.
Once you run consistent beads on flat plate, those habits transfer straight to fillets, laps, and out-of-position work. The guides on aluminum boat welding and alloy wheel welding show how much that matters on real parts.
Troubleshooting the Most Common Aluminum TIG Defects
Every aluminum TIG welder ends up as a detective: the bead looks fine, then porosity or a crack shows up later. Most aluminum defects are predictable and trace back to a short list of settings and habits. Here are the heaviest hitters with their cause and fix.
Porosity: The Default Failure Mode
Porosity almost never comes from the machine. Troubleshooting guidance from TWI on avoiding porosity when welding aluminium lists the documented causes: long arc length, high heat input, inadequate shielding, high humidity, oily material, and a high gas dew point. Most are prep or environmental problems. Hydrogen is the villain; it dissolves readily in molten aluminum, then comes out of solution as the puddle freezes, leaving bubbles trapped in the bead.
Shorten the arc, boost gas coverage if the cup is small for the joint, move indoors when humidity is high, and check your argon dew point if porosity returns. Excessive travel speed is a sneaky cause: rush the puddle and it freezes before dissolved gas escapes. If prep is the fault, run the solvent-then-brush routine from this guide’s prep section and our walkthrough on prepping aluminum for welding.
Floating Black Spots: Your Balance Setting Is Calling
Floating black spots drifting in the puddle are a satisfying quick win: your balance is too high in electrode negative. Too much EN means too little cleaning action, so oxides burn into black specks. Reduce the EN percentage until the puddle runs clear. Miller’s write-up on AC balance control explains it: more electrode positive buys oxide breakdown, at the cost of heat in the tungsten.
Quick Note: Black specks that float are a balance problem. Specks that stick to the tungsten and transfer are contamination. If reducing EN does not clear the puddle, check for a dipped tungsten or filler rod.
Lack of Fusion and Incomplete Penetration
Lack of fusion looks like a bead sitting on top of the joint rather than biting into both sides. The culprits are too little amperage, travel that outruns the puddle, or a torch angle pushing the arc onto one plate. Because aluminum pulls heat away so fast, a surface-hot bead can be cold at the root: add amps, slow down, and aim at the root.
Incomplete penetration is the same heat deficit; on thicker sections a bevel or gap gives the arc somewhere to bite. That heat-first mindset carries across processes, as our full guide to aluminum MIG welding shows.
Tungsten Contamination and Peppering
Contamination happens when the tip touches the puddle, filler rod, or a dirty edge, and the result is a wobbly arc, a wandering puddle, and tungsten flecks in the weld. The moment the arc acts strange, stop, regrind, and restart; powering through only spreads the damage.
Peppering, the tiny dark dots along the bead toes, usually means the electrode touched something or shielding failed. Check for drafts, a loose gas connection, or a cup too close to the work.
Arc Wander and Excessive Tungsten Erosion
An arc that refuses to stay put is usually geometry: a poorly ground or contaminated electrode, or a balance too aggressive in electrode positive. Regrind lengthwise and reduce EP time if the tip balls up.
Excessive erosion points to three causes: too much current for the electrode size, contamination, or excess electrode positive time. Match diameter to AC amperage and step up a size when running hot. The Fabricator on choosing tungsten type and size for aluminum GTAW covers that matching.
Cracking and Distortion
Hot cracking shows up in the crater or along a bead in a restrained joint. End every weld with downslope and post-flow so the crater fills and cools under shielding. 4043 and 4943 buy crack resistance, which is why they remain the go-to for many 6061 repairs.
Distortion is the quieter defect. Aluminum expands roughly twice as much as steel and shrinks about 6 percent during solidification, so long beads pull the workpiece out of shape. Fight it with balanced sequences, tacks placed to resist movement, and travel fast enough to avoid dumping heat in. Our aluminum boat hull welding guide shows why it matters.
Defect-to-Fix Lookup Table
| Defect | Likely Cause | Corrective Action |
|---|---|---|
| Porosity | Long arc, humidity, oily material | Shorten arc, 15–20 CFH, degrease |
| Floating black spots | Balance too high in EN | Reduce EN until clear |
| Lack of fusion | Low amps, fast travel, angle | Add amps, slow down, aim at root |
| Tungsten contamination | Tip touched puddle or filler | Stop, regrind, restart |
| Arc wander | Poor grind, contamination, high EP | Regrind lengthwise, reduce EP |
| Excessive erosion | Too much current, contamination, high EP | Use larger tungsten, reduce EP |
| Crater cracking | Abrupt arc stop | Downslope with post-flow |
| Distortion | Excess heat, long beads | Balanced sequence, tacks, speed |
Work down the table, change one variable at a time, and log what fixed the defect. That habit turns troubleshooting from guesswork into a repeatable procedure.
Safety, Ventilation and PPE for Aluminum TIG Work
Settings and hand skill only help if you can still see, breathe and feel your fingers after a few years at the bench. Aluminum GTAW carries a short list of serious hazards: intense ultraviolet light, ozone and metal fume, a surprisingly loud tone at high AC frequency, and cleaning chemistry that can turn a routine wipe-down into a toxic gas event. In U.S. shops the reference point is AWS Z49.1, Safety in Welding, Cutting, and Allied Processes, applied alongside OSHA’s hazard communication standard, 29 CFR 1910.1200. Z49.1 defines what protection the work requires; 1910.1200 requires that you actually know what is in the can you just opened.
Eye, Skin and Body Protection
The arc is the fastest route to injury. Ultraviolet output causes photokeratitis, commonly called arc eye, and the same radiation sunburns exposed skin on your neck, forearms and hands. Both injuries show up hours later, so the work feels harmless in the moment, which is exactly why protection gets skipped. Most AC TIG work sits around filter shade 10 to 13, and Z49.1 publishes filter shade tables that match shade to amperage. An auto-darkening helmet with a generous shade range saves you from swapping lenses between jobs, but keep clear safety glasses underneath for grinding and tacking, and hang a screen so bystanders are not taking UV from your arc. Miller’s beginner TIG aluminum series opens with the same baseline list: helmet with the correct filter, safety glasses, hearing protection and flame-resistant clothing.
Clothing in this process has one job: it must not catch and hold molten metal or sparks.
- Flame-resistant jacket or leathers, sleeves at the wrist, cuffs overlapping the glove tops.
- Fitted closed leather gloves, because bulky gloves cost you puddle control.
- High-top leather footwear with no exposed laces. Never sneakers.
- No polyester or fleece underneath. Synthetics melt and keep burning against skin.
- Hair tied back, no hoodies with drawstrings, no phone sitting in a pocket facing the arc.
Quick Note: Keep chlorinated solvents out of the welding area entirely. Trichloroethylene, perchloroethylene and chlorinated brake cleaner can off-gas near heat and UV, and the arc breaks them down into toxic gases such as phosgene. Degrease with a dedicated, low-residue product like the options covered in Aluminum Cleaner for Welding: How to Prep Aluminum Right, let the surface flash off completely, then weld.
Ventilation: What You Actually Breathe
Aluminum GTAW produces fine aluminum oxide fume plus ozone, and both accumulate fast in a closed garage. Position local exhaust close to the arc rather than relying on an open shop door, and treat any coating as a hazard until it is removed: anodizing, primers, chromate treatments, weld sealants, cutting fluids and shop oils all add their own chemistry to the plume. When fume cannot be controlled at the source, a properly fitted respirator is the fallback, selected under a written exposure plan and backed by safety data sheets, which is the documentation 1910.1200 exists to force. If you are new to aluminum prep work, our Aluminum Flux Welding: How It Works and When to Use It guide covers the same cleaning-before-arc principle from a different angle.
Electric Shock, Cylinders and Compressed Gas
High-frequency start and a wet floor are a bad combination. Inspect leads for cracked insulation, keep gloves dry, never weld standing in water or on a damp concrete pad, and make sure the work clamp sits on clean metal. Argon itself is inert, but it displaces oxygen silently, so a leaking cylinder in a small room can knock you out before you notice anything wrong. Secure cylinders upright with a chain or strap, fit the valve cap before moving one, open valves slowly, and never lift a cylinder by its cap. Liquid argon flashing to gas is cryogenically cold and causes instant frostbite burns to skin and eyes. Cylinder sizing, regulators and shop handling are covered in AllGas Welding Supply: Gases, Services & Safety Guide.
Noise: When AC Frequency Gets Loud
Frequency is a sound setting as much as an arc-shape setting. Everyday work runs at 100 to 120 Hz and stays reasonably quiet, but pushing AC frequency above roughly 250 Hz produces a loud, high-pitched tone that rings through the shop and into your head. The constricted arc is genuinely useful on thin material and fillets, so use it, but wear ear plugs or muffs whenever you do and warn anyone working nearby. Compressed air, grinders and cutoff wheels add to the daily dose.
| Hazard | Where it comes from | Practical control |
|---|---|---|
| Arc radiation | UV and visible output | Filter shade 10 to 13, safety glasses beneath, screens for bystanders |
| Fume and ozone | Oxide, coatings, oils | Local exhaust, strip coatings, respirator if source control fails |
| Phosgene | Chlorinated solvent near the arc | Use no chlorinated solvents in the welding area |
| Asphyxiation | Argon leak in a tight space | Ventilate, secure cylinders, monitor oxygen in confined work |
| Hearing loss | AC frequency above about 250 Hz | Ear plugs or muffs, warn people nearby |
| Shock | HF start, wet conditions, damaged leads | Dry gloves, inspected leads, grounded workpiece |
None of this is exotic gear. It is the same discipline you bring to cleaning and settings, applied to yourself. If you want structured instruction rather than self-teaching, a formal Aluminium Welding Course: What You Learn, Costs & Careers builds those safety habits into muscle memory, and shops working to code-level quality follow the process discipline described in Aerospace TIG Welding: Standards, Techniques, and Training. Get the protection right once and it stops being a decision you make each morning.
Practice Plan, Quick-Win Checklist and Final Tips
Knowing the settings is one thing. Reproducing them on a Tuesday afternoon with a deadline is another. The fastest way to turn everything in this guide into muscle memory is a short, staged practice plan that changes one variable at a time, plus a checklist you can tape inside the machine door.
A Five-Stage Progression That Builds Skill Deliberately
- Bead-on-plate at fixed settings. Start on 1/8-inch 6061 plate at 100 Hz, roughly 75% electrode negative, 15-20 CFH argon, and a comfortable amperage you hold steady. Ignore speed and focus only on arc length, a 5-15° torch tilt, and an even dab rhythm. Run ten beads. Judge them on consistent width and straight, evenly spaced ripples rather than how fast you finished.
- Balance and frequency experiments on scrap. Now change one variable per session. A typical documented AC balance test lays down four beads on the same plate at 90, 80, 70 and 60 percent EN so you can see the cleaning etch widen as electrode negative drops, then compare penetration and puddle viscosity side by side. Do the same at 80, 100 and 120 Hz. Photograph the results and label them; a Miller AC balance control walkthrough walks through the same comparison logic.
- Coupon welding on 1/8-inch 6061. One documented demonstration weld is a useful anchor point: 140 amps on a foot pedal, 3/32 lanthanated tungsten, 3/32 4043 filler, 100 Hz, and 15-20 CFH of argon. Butt coupons first, then cut and inspect the root for full fusion.
- Lap and fillet joints. These force you to aim the arc at the joint corner, manage heat in an asymmetrical shape, and add filler faster on the thick side. Run them in the flat position until the beads match, then move to vertical and overhead.
- Out-of-position work and real projects. Once overhead fillets hold up, apply the same discipline to a genuine job. If boats are your target, the techniques transfer directly to aluminum boat welding; if you are heading toward certification or shop work, the fundamentals here feed into formal aluminium welding course training.
Quick Note: Never evaluate a new balance or frequency setting by eye alone on the finished part. Run test welds on scrap of the same alloy and the same thickness you will actually weld, then clean, section or break them. Different thickness changes heat sink behavior enough to make a good setting look wrong.
Shop-Floor Quick-Win Checklist
- Degrease with a suitable solvent first, then stainless-wire brush with dedicated tools. Never reverse the order, and never use a chlorinated solvent near the arc.
- Weld within a short window of cleaning; stored aluminum grows hydrated oxide that drives hydrogen porosity.
- Use lanthanated or equivalent rare-earth tungsten on modern inverters, ground longitudinally at 15-30°, with a stickout matched to your cup.
- Start at 100-120 Hz, 75% EN, and amperage near the per-inch rule, then adjust with the pedal rather than the panel.
- Match filler to service: 4043 for crack and distortion resistance and anodizing aesthetics, 5356 for strength and ductility, and 5356 only on alloys carrying more than 2.5% magnesium.
- Terminate every bead with downslope and post-flow to avoid crater cracks.
- Keep filler clean and capped; wipe rod from the outer wrapper only when feeding it.
If your cleaning step keeps failing, a dedicated chemistry routine helps, as covered in aluminum cleaner for welding prep guidance.
What This Guidance Can and Cannot Do
Here is the honest limitation: every number in this article, including the 140-amp demonstration weld, is a documented starting point, not a welding procedure specification. Machine settings sheets are published to get you close, not to replace qualification. Before you commit a setting to production work, confirm it with practice welds on scrap of the same material and thickness, then inspect for fusion, porosity and cracking. If your work falls under a formal code, the procedure must be qualified properly, which is where resources such as AWS Welding Digest on GTAW and the AWS C5.5 filler guidance become your reference, not a forum post.
Aluminum rewards consistency more than cleverness. Clean the metal, control the arc, choose the filler for the service conditions, and let test welds verify every adjustment you make. That loop is the whole game. Master it, and the TIG welding aluminum tips in this guide stop being notes you reread and become habits that hold up on the first pass, every pass.
Frequently Asked Questions
What AC settings should I start with for TIG welding aluminum?
Start near 75% electrode negative for balance and 100-120 Hz for frequency, then adjust on scrap of the same alloy and thickness. Lower the balance percentage if you see black peppering in the puddle, and raise frequency when you need a tighter, more focused arc.
Which tungsten is best for TIG welding aluminum on an inverter machine?
2% lanthanated, ceriated or zirconiated tungsten are the practical all-around choices. Pure green-band tungsten is not recommended for inverters, and thoriated tungsten tends to form irregular balls and wander on AC.
Do I sharpen tungsten for AC aluminum welding?
Yes, most modern inverter guidance says to sharpen it and grind lengthwise at 15 to 30 degrees with a dedicated wheel. On older transformer machines the tip may ball up naturally, and some welders deliberately blunt it to resist melt-back at higher amperage.
Why does my aluminum TIG weld have porosity?
Porosity in aluminum is always caused by hydrogen, usually from oil, hydrated oxide, moisture or poor gas shielding. Solvent clean first, stainless wire brush second, check for gas line leaks and condensation, and avoid long arc lengths and excessive heat input.
What is the difference between 4043 and 5356 filler rod?
4043 resists cracking and distortion, welds brightly and handles sustained service above 150°F, while 5356 gives higher strength, better ductility and better anodized color match. Never use 5356 in长期 elevated-temperature service, and always use 5356 on 5xxx alloys with more than 2.5% magnesium.
Why is AC used instead of DC for aluminum?
Aluminum’s tough oxide layer melts far above the base metal, so AC’s positive half cycle blasts the oxide off while the negative half cycle provides penetration. DC electrode negative alone welds through the oxide poorly and can trap contaminants in the joint.
How do I stop tungsten contamination while learning?
Keep arc length roughly equal to the tungsten diameter, rest your hand on the table for stability, and never dip the electrode into the puddle or touch the filler rod. Clip off the contaminated section and regrind on a dedicated wheel before restarting.
How do I prevent crater cracks at the end of an aluminum weld?
Use the foot pedal or downslope control to taper the current before breaking the arc, and consider welding back into the bead rather than stopping abruptly. A proper post-flow keeps the crater shielded while it solidifies.
Nicholas Harper
Editorial TeamCovering welding processes, equipment selection, and shop-floor practice for WeldZen — with every claim checked against the governing code.
Related Studies
Get The Next Benchmark First
New field studies and calibration sheets, once a week. No spam.